*4.3. Limitations and Future Research Directions*

This study aims to optimize the PLES with the objective that all ecosystem services can be in surplus as far as possible and proposes corresponding optimizations at multiple scales. This study can provide a basis for decision-making on regional land use management and rational allocation of resources. However, there are some problems with this study. Firstly, in assessing ES, this study used several models, such as the RUSLE model, water balance equation, and CASA model, where differences in data sources and calculation methods can lead to differences in results. Although there are still no effective solutions to these problems, these methods are still widely used [54,55]. Additionally, due to the lack of data and the limitations of ecosystem service models, only the supply and demand of four ES were assessed, which is not comprehensive for the complete management of ES. More ES assessments should be added in future studies. In addition, the use of land-use types for the classification of PLES is a more straightforward method [56]. However, this approach ignores the complex multifunctionality of land. For example, arable land (paddy and dryland) is uniformly classified as production space without taking into account its ecological characteristics. Finally, the issue of scale is also one of the problems studied in this study, with spatial correlation results varying with unit size (grid cell or grain size) [26,57]. In this study, the identification of thresholds was based on the grid-scale using a hierarchical statistical approach. Random points, various grid cell sizes, and basin units should be selected in subsequent studies to explore the differences in the impact of PLES on the ES supply and demand imbalance.

#### **5. Conclusions**

Based on various models and methods, this study quantified the mismatch of supply and demand for the four ES in the Yellow River Basin and explores how the spatial pattern of PLES can be adjusted to keep the ES in supply and demand balance. The results show that in 2000, 2010, and 2018, the total supply of the three ecosystem services in the Yellow River Basin was greater than the total demand, except for carbon sequestration services. Along with the implementation of revegetation projects and the establishment of ecological reserves in the region, the supply of many ecosystem services was on the rise. However, increased urbanization and over-concentration of population and economy resulted in

a serious spatial mismatch between supply and demand for all four ecosystem services, especially in the major urban centers. The spatial mismatch in ES can be effectively reduced by optimizing the PLES, e.g., increasing the production spatial ratio can effectively increase the supply of grain production service and alleviate the contradiction between supply and demand of grain production service in certain regions. This study provides an optimization objective for the PLES optimization of other regions by providing an ideal region in the ternary phase diagram, i.e., one that can ensure that multiple ES are in surplus at the same time. The direction of optimization of other areas is determined by their relative position to the ideal area, and the amount of adjustment of PLES can be determined by the difference from the ideal area. The PLES optimization framework proposed in this study is very flexible, as reflected in the choice of ES and multi-scale optimization proposals, which can effectively reduce the deficit problem of regional ES in the process of practical application.

**Author Contributions:** Conceptualization, X.F.; data curation, X.W.; formal analysis, J.M.; funding acquisition, X.W.; investigation, X.F.; methodology, X.F. and J.Z.; resources, X.W.; supervision, X.W.; visualization, X.F., J.Z. and J.M.; writing—original draft, X.F. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research was funded by the National Key Research and Development Plan of China (2016YFC0501603), the Chinese Academic of Sciences, the Strategic Priority Research Program of the Chinese Academy of Sciences (XDA2002040201).

**Conflicts of Interest:** The authors declare no conflict of interest.
